A method of boundary determination in digital images of urothelial cells

作者:

Highlights:

摘要

Digital images acquired by high resolution scanning of urothelial cells are segmented into nuclear and cytoplasmic regions. A method which is a hybrid of a gradient tracking method and a thresholding method is used, and criteria are presented to automatically detect failures of the algorithm. Tests on a sample of benign and malignant urothelial cells show a significant improvement in both accuracy and reliability as compared with a single segmentation method.

论文关键词:Cytology automation,Urinary cytology,Edge detection,Scene segmentation,Cell images analysis,Boundary definition

论文评审过程:Received 22 December 1980, Available online 19 May 2003.

论文官网地址:https://doi.org/10.1016/0031-3203(81)90019-4